Open AccessSemi-automated registration and segmentation for gingival tissue volume measurement on 3D OCT images
Author(s) -
Geng Wang,
Nhan Le,
Xiaohui Hu,
Yuxuan Cheng,
Steven L. Jacques,
Hrebesh M. Subhash,
Ke Wang
Publication year - 2020
Publication title -
biomedical optics express
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.362
H-Index - 86
ISSN - 2156-7085
DOI - 10.1364/boe.396599
Subject(s) - optical coherence tomography , volume (thermodynamics) , image registration , segmentation , repeatability , biomedical engineering , point cloud , medicine , computer vision , artificial intelligence , computer science , dentistry , radiology , mathematics , statistics , physics , quantum mechanics , image (mathematics)
The change in gingival tissue volume may be used to indicate changes in gingival inflammation, which may be useful for the clinical assessment of gingival health. Properly quantifying gingival tissue volume requires a robust technique for accurate registration and segmentation of longitudinally captured 3-dimensional (3D) images. In this paper, a semi-automated registration and segmentation method for micrometer resolution measurement of gingival-tissue volume is proposed for 3D optical coherence tomography (OCT) imaging. For quantification, relative changes in gingiva tissue volume are measured based on changes in the gingiva surface height using the tooth surface as a reference. This report conducted repeatability tests on this method drawn from repeated scans in one patient, indicating an error of the point cloud registration method for oral OCT imaging is 63.08 ± 4.52µm (1σ), and the measurement error of the gingival tissue average thickness is -3.40 ± 21.85µm (1σ).